Publications

Wang, PJ; Li, X; Tang, JX; Yang, JY; Ma, YP; Wu, DR; Huo, ZG (2023). Determining the critical threshold of meteorological heat damage to tea plants based on MODIS LST products for tea planting areas in China. ECOLOGICAL INFORMATICS, 77, 102235.

Abstract
Tea (Camellia sinensis (L.) Kuntze) is a daily drink in modern life throughout the world. Global warming increases the exposure of tea plants to elevated heat conditions during the summer picking period. Therefore, it is a fundamental function of the meteorological service to determine the critical temperature threshold that triggers heat damage (HD) to tea plants. In this study, MODIS daytime land surface temperature (LST) products obtained from the Aqua satellite (MYD11A1), annual land cover type products (MCD12Q1), and actual HD records were jointly used to determine the critical LST threshold resulting in HD to tea planting areas in China. An LST dataset of HD and HD-free samples was constructed based on actual HD records. The box-plot method was then used to delineate the range of LST outliers in HD to tea plants. The cumulative frequency distribution (CFD) method was adopted to determine the critical LST threshold by searching for the maximum slope of CFD within the range of LST outliers based on 2901 actual HD samples from 42 agrometeorological stations in 11 provinces in China. Results showed that the critical LST threshold triggering HD to tea plants was 30.1 degrees C. Two different HD data sources (i.e., 68 counties in eight provinces from the Meteorological Disaster Management System (MDMS) and 37 counties in six cities from a published book) were used as HD validation samples. Validation accuracy analysis revealed that a critical LST threshold of 30.1 degrees C was reasonable, with an average accuracy of between 60% and 90% for the published-book dataset, and 75.7% in the northern Yangtze River area and 86.3% in the southern Yangtze River area for the MDMS dataset. This study confirmed that using remotely sensed LST products is an effective method for identifying temperature-related agrometeorological disasters owing to the method's ability to intuitively capture the surface temperature distribution of the disaster-affected object. The findings will be helpful for identifying the timing and location of HD to tea plants, and thereby potentially preventing HD to tea plants.

DOI:
10.1016/j.ecoinf.2023.102235

ISSN:
1878-0512